Evaluation of the New Deal for Lone Parents: early lessons from the Phase One prototype - synthesis report
A hard copy of this report summary can be obtained by contacting Paul Noakes [E-Mail: Paul.Noakes@dwp.gsi.gov.uk] or by writing to him at the 'Social Research Division, Department for Work and Pensions, 4th Floor, Adelphi, 1-11 John Adam Street, London WC2N 6HT'.
Research Report No. 108
By Jon Hales, Carli Lessof, Wendy Roth, Mandy Gloyer, Andrew Shaw Jane Millar, Matt Barnes, Peter Elias, Chris Hasluck, Abigail McKnight and Anne Green
This report presents findings from the evaluation of the New Deal for Lone Parents Phase One prototype which was commissioned by the Department of Social Security. It has been conducted by a consortium of independent researchers at the National Centre for Social Research (formerly SCPR), the Centre for Analysis of Social Policy at the University of Bath and the Institute for Employment Research at the University of Warwick.
The aim of the NDLP prototype (launched in July and August 1997 in eight areas across the UK) was to help lone parents on Income Support move into work or towards preparing for work through the key mechanism of the support of a personal adviser.
The evaluation incorporated a range of research components and this report aims to draw together the main findings across all the areas of research. A number of reports present findings from separate aspects of the evaluation: the Survey Report (Hales et al, 2000, DSS Research Report No. 109) is concerned with the findings of surveys conducted for the evaluation; the econometric and cost benefit analysis report consists of three papers on cost-benefit analysis and the pattern of transitions off and factors affecting Income Support (Hasluck, McKnight and Elias, Report 110) .
The main findings were:
- The NDLP prototype had a small positive effect on the number of lone parents moving off Income Support. After 18 months the number of lone parents on Income Support was 3.3 per cent lower than it would have been in the absence of the program.
- The evaluation estimates that 20 per centof jobs gained following participation in NDLP were additional to those that would have occurred without the program. About 80 per cent of those who participated in NDLP and moved into work would have found work anyway - i.e. they are 'deadweight'. Generally the programme was geared towards lone parents who were work-ready, rather than the 'harder to help'. Nevertheless 64 per cent of full participants said they had benefited from the programme.
- Nearly a quarter of target group lone parents came forward for interview following receipt of an introductory letter; just over a fifth of those who received a letter became full participants in the programme. Seventy seven per cent of target group lone parents did not respond to the letter.
- The overall cost of the programme per participant moving into work is estimated at £1,388
NDLP Phase One Prototype
The New Deal for Lone Parents was launched as a 'Phase One prototype' in July and August 1997 in eight areas across the UK, and was implemented nationally in October 1998. Like the national programme, the New Deal for Lone Parents Phase One Prototype was voluntary and aimed to help lone parents on Income Support move into work, or towards work. Personal advisers provided an integrated service of advice and support, covering job search, advice on training, help in finding childcare services, advice on benefits and help with claiming benefits. Although all lone parents on Income Support in the prototype areas were eligible, there was a 'target group'- those with children aged over five years and three months - who were sent invitation letters.
Objectives of the evaluation
The main objectives of the evaluation can be classified under four headings:
- “participation” (who participated and who did not participate in the programme, and for what reasons?)
- “lessons” (what did participants and advisers think was helpful in getting lone parents into work?)
- “resources” (what was the take-up among those eligible, what resources did they need and what additional demand was generated for other services?)
- “counterfactual” (how much movement into work was additional and could be attributed to the programme?).
A key feature of the evaluation was the use of six areas where the programme was not operating, to act as comparison areas. These were intended to show what the situation in the NDLP areas would have been in the absence of the programme.
Design of the evaluation
The evaluation incorporated a range of research components, including site visits, labour market studies, in-depth interviews, surveys of lone parents, analyses of administrative data, work and benefit histories, and an assessment of the costs and benefits of the prototype. These studies have been reported in a number of separate reports. The Synthesis Report aims to draw together the main findings across all the areas of research, both to provide an overview of the evaluation and also to provide pointers to where more information is contained in the set of more detailed reports.
Policy on lone parent families in Britain
The New Deal for Lone Parents should be seen as part of an increasingly significant policy focus on lone parent families. There have been striking increases in recent decades in the proportion of families headed by lone parents, and of those who are dependent on Income Support. Poverty and benefit dependency are features of many lone parent families, and this has been associated with their low levels of employment. While increasingly, other mothers have entered employment, lone mothers have faced particular barriers to the labour market.
Throughout the 1990s, there has been a gradual evolution of work incentive measures aimed at removing these barriers.
International experience
The New Deal for Lone Parents is one of a number of programmes in Britain which have been designed to encourage work among those who are able. Their emphasis on advisory services has similarities with programmes in other countries, for example the Australian Jobs, Employment and Training Scheme (JET) and the Californian Greater Avenues for Independence Program (GAIN). Both of these programmes have resulted in modest overall reductions in welfare benefit expenditure.
Voluntary and compulsory programmes in Britain and abroad
Participation in NDLP is voluntary and claimants can choose whether or not to take part with no implications for their entitlement to benefit. However, there has been a shift in most welfare to work policies from voluntarism towards the use of greater compulsion. This has been seen in Britain and abroad. While the clearest evidence of compulsion in Britain is associated with other client groups such as the unemployed, there is evidence of increasing conditionality in lone parents' receipt of benefit in the introduction of the ONE pilots, which started operation in four areas in June 1999 and in another eight areas from November 1999. In the ONE pilot areas, it will be compulsory from April 2000 for all clients of working age, including lone parents, to attend an advisory meeting at the start of a new or repeat claim for working-age benefits, including Income Support, and at other times when asked to do so.
Design of the prototype
Eight Benefits Agency districts constituted the Phase One prototype. Three levels of claimant unemployment were used in the selection of areas, with rates varying between 2.6 and 7.7 per cent at the start of the prototype(2). The size of the lone parent population claiming Income Support in each BA district ranged from approximately 5,000 to 13,000 and the number of advisers varied accordingly.
Six other Benefits Agency districts were selected as a basis for comparison to assist the evaluation, having been matched as far as possible on labour market characteristics, including rates of claimant unemployment.
Throughout the period of study - and in both the prototype and comparison areas - there was a fall in both unemployment rates and in the number of lone parents claiming Income Support. The percentages who stopped claiming Income Support were similar in the two sets of areas, although labour market conditions in the comparison areas were consistently more favourable.
Target Group
Lone parents whose youngest child was aged five or over were the target group for the programme, but others could put themselves forward. In the prototype areas, there were 33,332 members of the 'stock' target group at the outset of the programme, that is people who had Income Support claims which had started prior to May 1997(3). By the end of the prototype, there had been a much smaller number (5,700) of 'flow' target group lone parents in receipt of Income Support, those whose Income Support claim started in May 1997 or later.
Participation
Despite differences in the organisation of personal advisers between areas, the initial interview was a key aspect of the NDLP intervention in all areas. In order to encourage participation, advisers provided various forms of help, including a 'better off' calculation for those who were uncertain about the financial advantages of working, or working more hours. The advisers had to develop new approaches, deal with teething problems of new systems, and allay the suspicion of lone parents and community groups who sometimes feared there would be coercion into work regardless of the circumstances of the individual. After starting with much larger caseloads, advisers found that at any one time they managed caseloads of between 20 and 30 clients most effectively.
Movement into work
On average, each adviser helped a lone parent to start work every one and a half weeks. Those lone parents who were not invited to an interview but participated on their own initiative had the highest success rate in obtaining employment.
Cost per lone parent
The average unit cost of the prototype ranged from around £140 per lone parent invited to attend an interview to around £1,388 per lone parent who gained employment and left Income Support.
Effect of the prototype
The evaluation examined a number of measures of programme performance. These included movement off Income Support, movement into paid work, financial and other benefits associated with the programme.
Various sources of information were employed in the assessment of the effect of the Phase One prototype. These included administrative records of Income Support claims, labour market data, survey data collected by means of interviews with lone parents, site visits to meet advisers, two seminars with the adviser-managers of the eight areas, and financial data from the programme administration.
Comparing the monthly counts of Income Support claims by lone parents in the target group, the number of Income Support recipients fell in both the prototype and comparison areas. In October 1998, at which time the New Deal for Lone Parents was implemented nationally, the decline in the prototype areas was approximately one and a half percentage points greater than in the comparison areas.
Early and later invitation to participate
A second type of evidence on the programme's effect was based on a feature of the programme's implementation that made it possible to identify lone parents who were invited to participate in the programme earlier or later in the course of Phase One. Comparison of movements off Income Support for these two groups shows that by the time most of the later group of lone parents received their letters of invitation about six months after the start of the prototype, an additional one and a half per cent of the earlier group had moved off Income Support. Transitions off Income Support were modelled within an econometric framework using the data from administrative records, adjusting for lone parents' personal characteristics and local labour market conditions. On the basis of this analysis, we estimated that after eighteen months the stock of Income Support claims was 3.3 percentage points lower in the prototype areas than could have been expected if the programme had not existed.
Work and benefit history
The contrast between lone parents on Income Support in the comparison and prototype areas was also studied using the survey evidence. Statistical analysis of the work and benefit histories from the survey data, adjusting for local labour market conditions (represented by the female rate of claimant unemployment) and lone parents' characteristics, produced an estimate that by October 1998 the stock of Income Support claims was three percentage points lower in the prototype areas, although this was at a significance level of 93 per cent and so has not been treated as significant.
Movement off Income Support
In terms of their destinations, the survey conducted between mid-October 1998 and the end of January 1999 showed that 28 per cent of the sample of lone parents living in comparison areas had ceased to claim Income Support, compared with 30 per cent in the prototype areas. However, 17 per cent of lone parents in the prototype areas and 18 per cent in the comparison areas had moved into work. While these differences are not statistically significant, other evidence also suggests that a higher percentage of lone parents in the prototype areas were leaving Income Support and entering education or training, or had ceased to claim Income Support having re-partnered. Many of those in the prototype areas who had found work felt that their ability to start work had been influenced by their personal adviser.
Survey evidence also showed that lone parents who had started work reported that they felt they were financially better off in work than claiming Income Support.
Cost-benefit analysis
As well as the private benefit to the lone parent's household of having an income from work, increased employment could be beneficial for the economy and the Treasury. Evidence from the evaluation of costs and benefits suggests that the prototype programme resulted in economic returns which were slightly less than the cost of the prototype, representing a modest net cost to the Exchequer. This calculation was based on an estimate that about 20 per cent of the jobs gained by lone parents who participated in the programme were additional to the number of lone parents who could be expected to have started work in the absence of the programme. If the percentage of additional jobs had been 23 per cent, the programme would have had economic benefits equal to its costs.
While a financial assessment is important in terms of justifying the allocation of scarce resources to social programmes, it has to be recognised that with a programme such as the New Deal for Lone Parents there are many non-financial gains for lone parents, their children and society as a whole. Improving the living conditions of lone parent families may be seen as sufficient justification for the existence of the programme. An important potential gain, which cannot be quantified in the short term, concerns the longer-term benefits accruing to the children of participants in the New Deal for Lone Parents.
Personal advisers
The New Deal for Lone Parents prototype pioneered the use of personal advisers by the Benefits Agency and the Employment Service to deliver a Welfare to Work programme. The advisers were able to offer a wide range of information and advice, tailored to the personal situation of each client. However, there was some constraint in the limited size of each team of advisers, in relation to the size of the lone parent population. There was also a limited time available to demonstrate the effectiveness of the approach, which means that some of the benefits of personal adviser activity had not materialised by the end of 1998.
Initial and full participation
Participation rates, as indicated by the survey data, suggest that overall one in five (21 per cent) of the lone parents who were approached to take part in the Phase One prototype became full participants. This means that they had at least one interview with a personal adviser and discussed one or more of the topics which the programme was concerned to promote. A further three per cent had an initial interview with an adviser but did not discuss any of these topics or proceed further. Of all those who had an initial interview, 93 per cent are classified on this basis as full participants. Just over three-quarters (77 per cent) of the lone parents in the target group did not take part at all.
There was little difference between participants and non-participants in terms of demographic characteristics. But participants were likely to have less restricting family responsibilities (in terms of numbers and age of children in their household for example) and notably greater eligibility for the labour market in terms of qualifications and work experience. More were already looking for work. Proximity to the New Deal for Lone Parents office also affected take-up. However, of particular significance was the form of approach letter used. This greatly increased initial take-up and, ultimately, full participation in the programme, if it specified an appointment time, because this was often assumed by lone parents to indicate that attendance was compulsory.
Non-participants
Of the non-participants, 60 per cent actively decided that the programme was not appropriate for them, rather than it 'just ending up' that they did not participate. Of these, some were already in work (17 per cent of those who decided not to take part), or were in education (seven per cent), or on a training course (one per cent). Others said their children were too young (15 per cent), or their health precluded work (11 per cent) or were affected by circumstances at the time, such as illness (10 per cent) or other pressing matters (13 per cent). Eight per cent wanted to conduct their own jobsearch independently.
Advisory interviews
Initial advisory interviews varied appreciably in length. Participants on the whole had just one interview, usually with telephone follow-up and/or material sent by post. Occasionally they had a visit at home from the personal adviser. Generally, they gauged this as about the right amount of contact but some (21 per cent) would have liked more. Some lone parents expected to have further contact with a personal adviser after the end of the prototype (i.e. within the Phase Three programme).
The main issues discussed with advisers (for about two participants in three) were steps towards looking for a job, advice on benefits, and childcare. Four out of five participants obtained a 'better-off calculation', which in most cases (70 per cent) showed that they would be better off in work. Such information met with a mixed reaction. While half were pleased and surprised, others were disappointed that their in-work income would not be greater. The increased level of income did not necessarily lead them to consider paid work as in their own best interest.
A quarter of the participants received help with job applications or with a CV. Those who were given advice or help with seeking, applying for or deciding on a job, almost all said the adviser's help was good. About one half of the participants who started work and left Income Support felt that the programme had affected their jobsearch strategy by, for example, encouraging a more positive attitude to work or a more active and varied jobsearch.
Those participants who went on to start work occasionally received further support (e.g. with benefit claim forms, benefit problems, or childcare issues). There was also only a limited amount of referral to other services, such as Jobclubs or the Work-Based Training for Adults programme for unemployed people, either run by the Employment Service or other organisations. Advisers did not actively offer the reimbursements for travel or childcare costs that were available, so the use of these resources was very limited. As a result, we can draw few conclusions about the resources that might be needed if these services were taken up fully in a similar national programme.
Assessments of personal advisers
The role and manner of the personal advisers was pivotal in determining lone parents' overall assessment of the programme. A distinction was made between those advisers perceived as 'effective' or 'ineffective'. Views tended to polarise between these extremes, with four-fifths of participants taking the very favourable view. Interestingly, a good adviser was valued more for qualities that were supportive and understanding of the lone parent's situation (friendly, outgoing, positive, enthusiastic, relaxed, confident) than for practical assistance. Great benefit was attributed to having someone to talk to, 'on their side', who could provide information about options, and help make sense of the system. This was valued in the context of low self-confidence among many of the lone parents. Where helpful information was also obtained this was valued, though there were some criticisms regarding lack of information on childcare, job vacancies, and occasionally benefits.
Overall, lone parents very much valued the personal adviser approach. As a source of advice and support, advisers helped to build a more positive, confident attitude, a better sense of available options, knowledge and help regarding benefits, and knowledge of whether they would be better or worse off in work. Many lone parents felt they had been helped to do something that they might not have done alone. The personal adviser service was experienced as very different from the Benefits Agency's and the Employment Service's traditional stance towards customers.
As such, the approach may have accelerated movement towards paid work among a group of people who had been diffident in their ability to achieve this. But in many cases work would be in the longer term, and not necessarily achieved in the time available for the prototype. On the whole, the conclusion on the prototype is that personal advisers largely worked with people who were already on the way to starting work and who might have done so anyway.
From the prototype to a national programme
The report identifies some of the factors which make it inappropriate to extrapolate directly from the prototype to a national programme.
First, there were differences in the orientation of the prototype phase, compared with the national programme, which could work in either a positive or a negative way. Positive features of a prototype may be seen in the commitment and enthusiasm of staff, competition between teams, and urgency to do the job in the limited time available. Other than the short period of time available to the prototype, Phase One operated under three main constraints. These were the time of year when it began (the school summer holidays), the lack of institutional support arrangements for services to which lone parents could be referred for help in overcoming barriers to work, and the scale of the operation (in terms of the large numbers of lone parents who were potential participants). The national programme, however, is likely to see the development of partnerships between the ES and other organisations to deliver services, which should result in a programme well adapted to local needs.
Second, there may be scope for a national programme to provide a more broadly based range of options than existed during the prototype for those not ready to move rapidly into work. The prototype was often most effective when lone parents had already found a job and needed help with Family Credit, but it must be recognised that many of these were lone parents who would have found work anyway. For others, there was only limited evidence of the prototype encouraging training while continuing to claim Income Support.
Third, opportunities for providing broader support for lone parents do not appear to have been developed during the prototype. The national programme might draw upon complementary services that are external to the prototype programme. As examples of such services which may be suitable for lone parents we suggest that support groups could help lone parents overcome a lack of self-confidence about their ability to work, and work intermediaries might provide support in the transition to work, perhaps as agency work rather than mainstream employment.
Fourth, the national programme now operates in a different setting to Phase One. Since the prototype was introduced, other policy developments have led to complementary programmes that improve access to childcare and increase the level of in-work financial support. The “National Childcare Strategy” (DfEE, 1998) is a significant change that should support both part-time and full-time work. Initiatives that make work financially more attractive include the minimum wage and, from October 1999, the “Working Families Tax Credit”, (including a childcare tax credit for those using registered childcare services), which further raises the levels of support available at the time of the prototype through Family Credit.
Estimating the early impact of the programme
The evaluation's overall conclusion is that the Phase One New Deal for Lone Parents had a small but appreciable effect on the rate of movement off Income Support and into work among lone parents in the eight Benefits Agency districts where it was implemented. This conclusion is supported by evidence of a fall in the number of lone parents claiming Income Support, which was greater in the eight prototype areas than in the six comparison areas. This finding is supported by multivariate techniques that were applied to administrative and survey data. In particular, this small but significant New Deal for Lone Parents effect was demonstrated by a greater likelihood to stop claiming Income Support among members of the stock, target group who were invited early in the programme.
Costs and benefits of the prototype
While it is difficult to estimate the net costs of the Phase One prototype with any precision, we estimate that there are significant social benefits consistent with the policy that were achieved at a relatively small Exchequer cost (just 12 per cent of the direct expenditure on the programme). Only a small change in one of the key parameters that determine costs and benefits would bring the net Exchequer cost closer to or beyond the break-even point.
Positive outcomes of the prototype
A number of positive outcomes of the prototype were identified. Almost half of those who participated were successful in finding jobs during the period in which the prototype operated and most were impressed by efforts made by personal advisers. A quarter of lone parents (28 per cent) who started work said their personal adviser had given them significant help in achieving this. The nature of this help was in boosting confidence and encouraging a positive attitude, rather than identifying vacancies and acting as an advocate with the employer.
Constraints on the prototype
The most serious constraint that the prototype faced was probably the focus on providing a service for as many as possible. As a result, advisers devoted the majority of their resources to lone parents who came forward most readily and were already preparing to look for work. While this had many positive benefits, fewer resources were then available to encourage those who faced greater barriers to returning to work (and who might be harder to reach). In part, this was necessitated by the scarcity of other resources such as job assistance schemes and training that the personal advisers could call on.
Complementary policy developments
In conclusion, it can be argued that the prototype gave sufficient evidence that a voluntary personal adviser service was workable and demonstrated positive results that justify the introduction of a national programme. Two key facilitators of a future welfare to work programme for lone parents are the National Childcare Strategy and the Working Families Tax Credit. These initiatives complement Phase Three of the New Deal for Lone Parents, while the existence of the programme can do much to ensure that the new measures are understood among the lone parents claiming Income Support, which is likely to be a critical factor in their impact.
Relevant publications
Elias, P. (2000). 'Transitions off Income Support: estimating the impact of the New Deal for Lone Parents using survey data' in Hasluck et al (2000) “Evaluation of the New Deal for Lone Parents: Early Lessons from the Phase One Prototype” - Cost-Benefit and Econometric Analyses, DSS Research Report No. 110, CDS: Leeds.
Finch, H., O'Connor. with Millar, J., Hales, J., Shaw, A. and Roth, W. (1999) “New Deal for Lone Parents: learning from the protoype areas”, DSS Research Report No.92, CDS: Leeds.
Green, A. (2000) “Evaluation of the” “New Deal for Lone Parents: A comparative analysis of the local study areas, ”DSS Social Research Branch, In-house Report No.63.
Hales, J., Shaw, A. and Roth, W. (1998) “Evaluation of the New Deal for Lone Parents: A Preliminary Estimate of the Counterfactual, ”DSS Social Research Branch, In-house Report No. 42.
Hales, J., Roth, W., Barnes, M., Millar, J., Lessof, C., Gloyer, M. and Shaw, A. (2000b) “Evaluation of the New Deal for Lone Parents: Early Lessons from the Phase One Prototype - Findings of Surveys”, DSS Research Report No. 109, CDS: Leeds.
Hasluck, C. (2000) 'The Net Economic and Exchequer Benefits of the New Deal for Lone Parents' in Hasluck et al (2000) “Evaluation of the New Deal for Lone Parents: Early Lessons from the Phase One Prototype” - “Cost-Benefit and Econometric Analyses”, DSS Research Report No. 110, CDS: Leeds.
McKnight, A. (2000). 'Transitions off Income Support: estimating the impact of the New Deal for Lone Parents using administrative data' in Hasluck et al (2000) “Evaluation of the New Deal for Lone Parents: Early Lessons from the Phase One Prototype” - “Cost-Benefit and Econometric Analyses”, DSS Research Report No. 110, CDS: Leeds.